Training distributed deep recurrent neural networks with mixed precision on GPU clusters.
Alexey SvyatkovskiyJulian Kates-HarbeckWilliam TangPublished in: MLHPC@SC (2017)
Keyphrases
- recurrent neural networks
- recurrent networks
- echo state networks
- feedforward neural networks
- neural network
- feed forward
- complex valued
- distributed systems
- long short term memory
- hidden layer
- clustering algorithm
- reservoir computing
- feed forward neural networks
- artificial neural networks
- high precision
- real time
- peer to peer
- data transfer
- neural model
- multi agent
- training process
- cascade correlation
- fuzzy clustering
- parallel implementation
- deep learning
- training set
- neural models
- nonlinear dynamic systems
- deep architectures
- control algorithm